The integumental profile: A reflection of the underlying skeletal configuration?

The integumental profile: A reflection of the underlying skeletal configuration? M. H. Kuyl, DDS,· R. M. H. Verbeeck, PhD SCi,b and L. R. Dermaut, DDS, PhDc Gent, Belgium The aim of this study was twofold: (1) to evaluate the importance of the level of training in orthodontics when estimating skeletal configuration by visual inspection of the soft tissue profile, and (2) to evaluate a possible discrepancy between integumental profile (IP) and skeletal class (SC). Four test groups comprising 10 orthodontists (0), 10 senior assistants (S), 10 junior assistants (J), and 10 dentists (0) assessed horizontal and vertical skeletal pattern from a series of slides of 100 patients. The assessments were repeated after a 1-month interval. Cephalometric analysis was also carried out by using a number of conventional analyses. Analysis of the results with Levene's test, two factor mixed-design variance analysis, and Newman-Keuls' Multiple-Range test showed that (1) orthodontists, independent of their level of training, are more consistent in assessing an IP than dentists; (2) assessments were more consistent for sagittal profile than for vertical profile; (3) sagittally the Wits' appraisal corresponds best with IP, and (4) vertically the Steiner analysis corresponds best with IP. (AM J ORTHOD OENTOFAC ORTHOP 1994;106:597-604.)

Cephalometries has been used in orthodontics for a long period of time. Besides its application for diagnostic purposes, it offers information about treatment effects and is a major help in evaluating growth. There is, however, some debate concerning the necessity of cephalometries for evaluating a patient's growth pattern, some authors claiming that well-trained orthodontists are able to deduce skeletal profile by careful inspection of the soft tissue profile. Others claim to the contrary, that cephalometries is indispensable to evaluate a patient's growth pattern and doubt that soft tissue profile always reflects skeletal profile. It therefore remains uncertain which information is more relevant to arrive at the best treatment plan; perhaps we should pay more attention to the soft tissue profile instead of the underlying skeletal pattern when proposing treatment objectives? Many well-known names in the orthodontic field have given their opinions concerning the relationship between the soft tissue integumental profile and the underlying skeletal structures (Angle, I Broadbent," Brodie;' Tweed," Bjork," Riedel," Burstone," Downs," Subtelny,? Zachrisson et al."'). Not surprisingly there is no concensus, 'Senior resident. "Head of the department of dental matenals, University Gent. Belgium; Research Associate of the N.F.S.R. (Belgium). 'Head of the department of orthodontics University Gent. Belgium. Copyright © 1994 by the Amencan ASSOCIatIOn of Orthodonnsts, 0889-5406/94/$3.00 + 0 8/1/46913

although there is a tendency for orthodontists and oral surgeons to rely more on the clinical impression than cephalometric analysis when planning treatment of individual cases (Wylie."). In other words, while regarding cephalometries as a convenient source of information about growth and the growth potential of the patient, orthodontists and oral surgeons tend to look at the soft tissue profile first. Evaluation of the soft tissues may be at least as important as cephalometric analysis (De Smit, Dermaut':'). In 1990 Michiels and Tourne" concluded that there are no published studies that quantify the clinical performance of cephalometric measurement. This study attempts to determine whether the level of training in orthodontics is significant when estimating underlying skeletal configuration by visual inspection of the soft tissues, and to evaluate the possible discrepancy between integumental profile (IP) and skeletal class (SC). MATERIALS AND METHOD The sample

One hundred patients were randomly selected from the patient files of the orthodontic department of the University of Gent. The only criterion for selection was the presence of a pretreatment cephalogram and a good set of slides. The severity of the skeletal discrepancy was not considered during selection, although six cases of severe growth discrepancy were present in the sample (three Class II and three Class III cases). 597

Amencan Journal of Orthodontics and Dentojactal Onhopedu \ December Ilj
598 Kuyl, Verbeeck, and Dermaut

Table I. Age-analysis of the randomly selected group of patients for this study Sex

I

I

n

100 87 «18yr.) 13 (>18yr.) 55 45

Mixed Females Males

I

Age-distribution

Mean age

8-50 yr.

14

-I. 2

8-50 yr. 8-28 yr.

14,6 13,5

3. 1

4. 9

Table II. Distribution of anteroposterior growth pattern according to Sassouni, Steiner, and Wits

and the vertical growth pattern according to Sassouni, Steiner, and the y-axis (Bjork) (N = normal, 0 = open, D = deep) Anteroposterior growth Class I

Sassouni Steiner Wits/Bjork

n = 24 n = 41 n = 15

I

Class II

n = 56 n = 54 n = 62

I

Class III

Vertical growth

I

n = 20 n = 5 n = 23

Table I shows an age analysis of the sample; of the 100 patients, 87 were young adults; the male-female distribution was 45/55. Cephalometries

The cephalograms were traced and analyzed by using several analysis methods, all by the same orthodontist, and 180 measurements were repeated after a long time interval. The intraoperator cephalometric technique error was calculated to be 4%. For sagittal skeletal pattern, the analyses according to Steiner," Wits," and Sassouni 14 were used, and for vertical skeletal pattern those of Steiner," Sassouni," and the y-axis of Bjork." Table II shows the distribution of horizontal classification and vertical growth pattern given by the different analyses. Remarkably wide variation exists between the results obtained from the various analysis methods. Soft tissue assessment

To assess soft tissue profile, the slides of the 100 cases were shown to four groups of 10 persons each. Group 0 consisted of well-trained orthodontists, group S of senior postgraduate students (fourth year of their orthodontic training), group J of junior postgraduate students (second year of their orthodontic training), and group D of dentists of several years experience but with no additional orthodontic training. Each set of slides relating to one patient consisted of four views (profile, full face in relaxation, full face smiling, and 45°) and were numbered 1 to 100. The sets were stored in slide trays in random order and were projected for each group of assessors twice, with a minimum interval of 1 month between screenings to avoid bias. At both screenings the assessors were asked

Total

N

n = 100 n = 100 n = 100

n = 26 n = 19 n = 21

I

0 n = 21 n = 65 n = 30

I

D n = 52 n = 16 n = 49

I

Total

n = 100 n = 100 n = 100

to deduce the horizontal and vertical skeletal pattern of each patient and to note this on a form. Sagittal relationship was scored as Class I, II, or III, whereas vertical relationship was evaluated as "normal," "open," or "deep," as described by Sassouni." In this way two sets of 100 skeletal diagnoses were obtained from each assessor. Statistical methods

The statistical evaluation was carried out by using Levene's test and an analysis of variance (ANOYA) for the detection of inhomogeneities in the variance and of significant differences between the scores of the groups. A Newman-Keuls' multiple range test was used for ranking the groups. RESULTS

As shown by Table II, there were quite obvious differences in the results given by the various cephalometric analyses for skeletal pattern. Anteroposteriorly the Sassouni and Wits' results are fairly comparable, but the Steiner, analysis shows a significant shift from Class III toward Class I. The same holds true for the measurement of vertical growth pattern where the shift is noticeable in the Steiner analysis from "deep" toward "open." Repetition test (Duple-test)

The purpose of this test is to evaluate the capability of the different test groups to score the integumental profile (IP) consistently (i.e., repeating the same score irrespective of the underlying skeletal configuration).

Amencan Journal of Orthodontics and Dentofacial Orthopedics Volume 106, No.6

Kuyl, Verbeeck, and Dermaut

599

Sagittal Profile 100% re

ated score

75

75

50

50

25

25

o

o

s

o

o

J

o

s

J

o

Fig. 1. Number of consistent scores for sagittal and vertical IP (integumental profile) obtained by each testperson as a function of his training.

Mean Scores 100 % repeated score 80 Profile

..

60

CJ

40 20 0

0

S

J

Sagittal Vertical

D

Fig. 2. Global consistency of sagittal and vertical IP score as a function of training.

Fig. 1 summarizes the number of repeated (i.e., consistent) scores for the sagittal and vertical IP obtained by each test person as a function of his training. According to Levene's test the variance of the repeated score for the test groups considered is homogeneous (0.05 < p < 0.10) independent of the IP considered, so that the standard deviation "s" of the individual score of the test persons can be assumed as s = 7. The mean of the repeated score for the different test groups is represented in Fig. 2 as a function of the IP. Fig. 2 suggests in the first place that the consistency of the IP score increases with the degree of training. Moreover the consistency of the sagittal IP scores apparently is systematically higher than that of the vertical IP scores. This is confirmed by an ANOVA analysis based

on a two-factor mixed-design and the data represented in Fig. 1. The results of this analysis are summarized in Table III. The profile (sagittal-vertical) and the training (O-S-J-D) were taken as factors. Correspondence between skeletal class and integumental profile

As the first diagnosis generally forms the basis for the orthodontic treatment, only the set of scores obtained during the first session were used further to investigate the correspondence correlation between the IP and the Sc. Fig. 3 gives an illustration of the number of times that a sagittal IP of a patient corresponds with the SC according to the criteria of the Sassouni," Steiner," and Wits" analysis and was based

600

Amencan Journal of Orthodonucs and Dentofuciul Orthopedic\ December 199~

Kuyl, Verbeeck, and Dermaut

Sassouni

Steiner

Score

Score

;~~ ;~~ o

20

40 60 80 Patient nr.

100

o o

20

40 60 80 Patient nr.

100

;~o

o

20

40 60 80 Patient nr.

100

Fig. 3. Illustration of number of times a sagittal IP of a patient corresponds with SC (skeletal classification) according to Sassouni, Steiner, and Wits based on results of all test persons disregarding their training.

Table III. Results of the ANOVA according to a two-factor mixed design with the data in Fig. 2 Source

I

Profile (P) Training (T) Subject (per T) PxT Subject x P (per T) TOTAL

ss 3213 1613 2423 37.7 1569 8856

I

df

1 3 36 3 36 79

I

MS

3213 538 67.3 12.6 43.6

I

F

P

73.74 7.99

0.0000' 0.0005'

0.29

0.8351

'Statistically significant difference.

on the results of all test persons regardless of their level of training. In the same way the correspondence between the vertical IP and the SC according to the Sassouni," Steiner," and Bjork" analysis is illustrated in Fig. 4. Figs. 3 and 4 clearly demonstrate the enormous variation in the correspondence between the observed IP and the SC based on different classification. However, when one considers the mean score that a correspondence is found between the IP and the SC as a function of the degree of training and of the different classification criteria, one obtains Fig. 5. According to the Levene's test, the variance of the correspondency scores of the individual test persons with a certain degree of training and based on the different criteria for the sagittal IP are homogeneous (p > 0.20). The mean standard deviation of the correspon-

dency scores for the groups considered in Fig. 5, A can then be estimated as Ssag = 4.4. Table IV summarizes the results of an ANOVA-analysis by using a two-factor mixed-design and the individual scores on which Fig. 5. A is based. The training level (O-S-l-D) and SC (Sassouni. Steiner, Wits) were taken as the factors. According to Table IV, the interaction between SC and training level of the test persons is insignificant. This indicates that the difference in the corresponding scores of the test persons from the groups 0, S, 1, and D is independent of the underlying Sc. However, Table IV shows clearly that the degree of correspondence between IP and SC differs significantly between the test groups with a different training level and between the criteria used for the selection of the Sc.

Amencan Journal of Onhodonttcs and Dentofacial Orthopedics Volume 106, No. 6

Kuyl, Verbeeck, and Dermaut

SasSOlIli

;~~ 20

40

60

4._ Steiner

Score

o

601

80

100

Patient nr.

30

.

~~~ o

20

40

60

80

100

Patient nr.

;o

20

40

60

80

100

Patient nr.

Fig. 4. Illustration of number of times a vertical IP of a patient corresponds with SC (skeletal classification) according to Sassouni, Steiner, and Bjork based on results of all test persons disregarding their level of training.

The Newman-Keuls' Multiple-Range test on the 95% confidence level shows that the correspondence between sagittal IP and SC decreases according to = S = J > D. The same test indicates that the correspondence between sagittal IP and SC is significantly different when using the criteria of Wits' and Steiner, but not when the criteria of Sassouni and Steiner or those of Sassouni and Wits are used. On the basis of these results and Fig. 5, A, one can conclude that the Wits' analysis corresponds best with the IP scored by the test persons. Considering the Wits' appraisal, the mean correspondency score for orthodontists (i.e., the pooled groups 0, S, and J) then amounts to 63% as compared with 57% for group D. According to Levene's test, the variance of the correspondency scores of the individual test persons with a certain degree of training and based on the different criteria for the vertical IP, are homogeneous (p > 0.20). The mean standard deviation of the correspondency scores for the groups considered in Fig. 5, B can then be estimated as Svert = 6.5. Since the interaction between SC and training is not significant according to Table V, the difference in correspondency scores of group 0, S, J, or D does not change significantly with the different criteria used for defining the SC. Moreover, Table V shows that the mean correspondency scores that

°

are based on the pooled scores of the SC differ insignificantly between group 0, S, J, and D. However, the difference between the mean correspondency scores that are based on the pooled scores of group 0, S, J, and D is significant for the SC considered. A Newman-Keuls' Multiple-Range test on the 95% confidence level proves that the correspondence between the observed vertical IP and SC decreases in the order: Steiner > Sassouni = Bjork. On the basis of these results and Fig. 5, B, one can conclude that the Steiner standards were generally best recognized by the test persons. However, the global mean correspondency score between the vertical IP and the Steiner SC that are based on the scores of all groups and amounting to 43.2% is drastically lower than that found for the sagittal IP and the Wits' Sc. DISCUSSION

In this study, different cephalometric analyses on the same patient showed a remarkable difference in diagnosis according to the horizontal skeletal classification and the evaluation of the vertical growth pattern. The interpretation of different cephalometric standards on the same patient in an attempt to diagnose a skeletal discrepancy remains a matter of debate. According to the "repetitiontest," some patients were scored poorly by each of

Amencan Journal

30 L--_----...,--.-~----' Sassouni elner Bjork Fig. 5. Mean score for correspondence between IP and SC as a function of level of training and of different classification criteria.

the groups, whereas some skeletal profiles were very well recognized. It was obvious that sagittal discrepancies were usually better scored than vertical ones. The consistency of scoring the sagittal discrepancies for all groups together was found in 75% of the patients, whereas the vertical discrepancies were scored identical in 63% of the cases. Moreover, the better trained test persons apparently are more accurate in the "repetition test," according to Fig. 2. However, the appropriate statistical test reveals that there is no significant difference in the repetition test between the groups of orthodontists, i.e., group 0, S, and J. In other words, the degree of training in these three groups was found to be of no significance with respect to

0/ Onhodontu.s

and Dentofacial Orthopedic; December 1'194

reliability in the repetition test. These three groups. however, were scoring more consistently than the dentists. The recognition of the underlying skeletal discrepancy by evaluating the soft tissue profile is apparently not easy. According to Fig. 5, there was only a minor training effect in this matter. The same conclusion seems to hold true as in the repetition test. No significant differences were found between orthodontists and orthodontists under training (i.e., groups 0, S, and J). Again, the dentists were recognizing to a lesser extent the underlying skeletal profile. The vertical profile score was in general lower than the sagittal profile score regardless of the cephalometric analysis used. The sagittal profile score was found to be the same when the Sassouni's analysis and the Steiner analysis were used for all test groups. Only a small difference was noted between the Wits' appraisal and the Steiner standards. With reference to the Wits' appraisal, dentists with advanced orthodontic training were able to recognize the underlying sagittal discrepancy in 63% of the patients. This was the case in 57% of the patients for dentists. There were no significant differences for the vertical profile scores between the different test groups. No effect of training at all was found for this variable. Dentists were as able as orthodontists to score the vertical profile. However, this score was rather low for all the groups. When using the standards from Bjork and Sassouni, no differences were found between the two for the vertical profile score. The Steiner norms for vertical discrepancies were to some extent better recognized by evaluating the soft tissue profile. CONCLUSIONS

1. The repetition test (duplo-score) indicates

that sagittal discrepancies were scored more reliably (75%) than the vertical discrepancies (63%). Both scores, however, show a considerable error indicating that the consistency in scoring an underlying skeletal discrepancy by evaluating the soft tissue profile is not high. 2. Dentists were scoring less consistently than orthodontists in the repetition test. The level of advanced training in orthodontics, however, had no influence on the consistency of scoring. 3. The recognition of the underlying skeletal

Amencan Journal of Orthodonucs and Dentofactal Orthopedics Volume 106, No. 6

Kuyl, Verbeeck, and Dennaut 603

Table IV. Results of the ANOVA using a two-factor mixed-design and the individual correspondency

scores on which Fig. 5, A is based Source

SS

df

MS

F

P

SC Training (T) Subject (per T) SC x T Subject x SC (per T) TOTAL

146.6 602.7 1043 119.6 1061 2972

2 3 36 6 72 119

73.3 200.9 29.0 19.9 14.7

4.98 6.94

0.0095' 0.0011*

1.35

0.2445

*Statistically significant difference.

Table V. Results of the ANOVA using a two-factor mixed-design and the individual correspondency

scores on which Fig. 5, B is based Source

I---S-S---

df

MS

F

P

Sc Training (T) Subject (per T) SC x T Subject x SC (per T) TOTAL

679.5 179.3 1582 304.9 3028 5774

2 3 36 6 72 119

339.7 59.8 44.0 50.8 42.1

8.08 1.36

0.0010* 0.2698

1.21

0.3114

*Statistically significant difference.

discrepancy by evaluating the soft tissue profile was found to be not easy. Again, dentists had more problems to do so than dentists with advanced-orthodontic training (orthodontists and orthodontists under training). 4. The sagittal profile score was not found to be different between Sassouni's analysis and the Steiner analysis for all test groups. Only a small difference was noticed between the Wits' appraisal and the Steiner standards. The best reference was the Wits' appraisal reflecting in 65% agreement between the soft tissue profile and the underlying skeletal growth pattern for "dentists with advanced orthodontic training." A percentage of 57% was found for the group of "dentists." 5. The highest reflection of the underlying vertical skeletal pattern, evaluated by the soft tissue profile, was found by using the Steiner analysis. The Sassouni-standards and the yaxis scored equally. It was found that the vertical scores were much lower than the sagittal ones. Differences between the four different test groups were negligible indicating that advanced training was not contributing to an improvement of the vertical score. 6. From this study, one can conclude that the

soft tissue profile does not reflect the underlying skeletal growth pattern very well. This does, however, not mean that cephalograms are more accurate in determining the final treatment plan. An additional conclusion could be that some more significance should be attached to soft tissue profile evaluation than to cephalometric analysis in orthodontic diagnosis and treatment planning. It was not, however, the purpose of this study to find an answer to this complicated question. We thank Prof. Dr A. De Smit and his staff, Department of Orthodontics, V.u.E. (Brussels), Prof. Dr H. De Clerck and his staff, Department of Orthodontics, V.c.L. (Brussels), Prof. Dr L. Martens and his staff, Department of Pedodontics, R.V.G. (Gent), and Dr N.E.P. Bennitt for their cooperation in this research project.

REFERENCES 1. Angle EH. Malocclusion of the teeth. 7th ed. Philadelphia: S.S. White Dental, 1907. 2. Broadbent HB. The face of the normal child. Angle Orthod 1937;7:183-208. 3. Brodie AG. On the growth of the human head from the third month to the eighth year of life. Am J Anat 1941;68: 209-16. 4. Tweed CH. The Frankfurt mandibular plane angle in orth-

604

5. 6. 7. 8. 9.

10.

11.

Kuyl, Verbeeck, and Dermaut

odontic diagnosis. Classification, treatment planning and prognosis. AM J ORTHOD ORAL SURG 1946;32:175-204. Bjork A. The face in profile. Lund: Berlingska Boktryckeriet, 1947. Riedel RA. Esthetics and its relation to orthodontic therapy. Angle Orthod 1950;20:168-78. Burstone CJ. The integumental profile. Angle Orthod 1959; 29:93-104. Downs WB. Analysis of the dentofacial profile. Angle Orthod 1956;26:191-212. Subtelny JD. A longitudinal study of soft tissue facial structures and their profile characteristics defined in relation to underlying skeletal structures. AM J ORTHOD 1959; 45:481-507. Hillesund E, Fjeld D, Zachrisson BU. Reliability of softtissue profile in cephalometries. AM J ORTHOD 1978;74:53750. Wylie GA, Fish LC, Epker BN. Cephalometries: a comparison of five analyses currently used in the diagnosis of dentofacial deformities. Int J Adult Orthod Orthognath Surg 1987;2:15-36.

American Journal of Orthodontics and Dentojacial Orthopedics December 1994

12. De Smit A, Dermaut LR. Soft-tissue profile preference. AM J ORTHOD 1984;86:67-73. 13. Michiels LYF, Tourne L. Nasion tube vertical: a proposed method for testing the clinical validity of cephalometric measurements applied to a new cephalometric reference line. Int J Adult Orthod Orthognath Surg 1990;5(1):43-52. 14. Sassouni V. Archial analysis in three dimensions, AM J ORTHOD 1958;44:433-63. 15. Steiner Cc. Cephalometries for you and me. AM J ORTHOD 1953;39:729-55. 16. Jacobson A. The "Wits" appraisal of jaw disharmony. AM J ORTHOD 1975;67:125-38. Reprint requests to: Dr. L. R. Dermaut

Department of Orthodontics University of Gent De Pintelaan 185 B-9000 Gent Belgium

AAO MEETING CALENDAR

1995 1996 1997 1998 1999 2000

San Francisco, Calif., May 13 to 18, Moscone Convention Center (International Orthodontic Congress) Denver, Colo., May 11 to 15, Colorado Convention Center Philadelphia, Pa., May 3 to 7, Philadelphia Convention Center Dallas, Texas, May 16 to 20, Dallas Convention Center San Diego, Calif., May 15 to 19, San Diego Convention Center - Chicago, III., April 29 to May 3, McCormick Place Convention Center